An empathic GPT-based chatbot to talk about mental disorders with Spanish teenagers

TL;DR

This work presents a Telegram-based empathic chatbot for Spanish teenagers to discuss mental disorders using a semi-controlled, self-disclosure–driven approach. The system combines psychologist-designed prompts with open-ended dialogue powered by GPT-3 (via translation) to foster empathetic engagement while safeguarding user privacy and safety. Empirical results show meaningful user engagement, language-usage signals distinguishing indicated from healthy participants, and generally positive user experience, though translation gaps and model limitations were observed. The study demonstrates feasibility and invites future work with native Spanish-language models, retrieval-augmented generation, and multimodal capabilities to enhance safety, relevance, and impact in teen mental-health support.

Abstract

This paper presents a chatbot-based system to engage young Spanish people in the awareness of certain mental disorders through a self-disclosure technique. The study was carried out in a population of teenagers aged between 12 and 18 years. The dialogue engine mixes closed and open conversations, so certain controlled messages are sent to focus the chat on a specific disorder, which will change over time. Once a set of trial questions is answered, the system can initiate the conversation on the disorder under the focus according to the user's sensibility to that disorder, in an attempt to establish a more empathetic communication. Then, an open conversation based on the GPT-3 language model is initiated, allowing the user to express themselves with more freedom. The results show that these systems are of interest to young people and could help them become aware of certain mental disorders.

Figures (8)

• Figure 1: Percentage of users interviewed for each age group.
• Figure 2: Components on the dialogue system
• Figure 3: The chatbot conversation flowchart.
• Figure 4: Distribution of users and messages over time.
• Figure 5: Number of interactions according to the mental disorder and user type.